This essay focuses on the differences in the allocation of time across the skill distribution of the US population between 1965 and 2003. The population is segmented in 2 skill groups: individuals with educational attainment below or equal to 12th grade, and individuals with higher schooling. We find that the increase in leisure for the overall US population holds for both low and high skill throughout the whole sample period. Even though high skilled spent more time in leisure activities in 1965, this group ended up with an average 5 hours below than the one for the low skilled. This was a consequence of the higher increase in leisure time by the low skilled relative to the high skilled. This pattern was even more pronounced since 1985.

To understand these differences, we further disaggregate the population by sex, work status, and age. Two results stand out. First, the trends in leisure are robust across the different demographics we have considered. That rules out compositional changes as the main explanation for the asymmetric increase in leisure. Secondly, for males, the market behavior of hours across skills mirrors the one for leisure. The key explanation for the cross-sectional trends in leisure for men is therefore the different behavior that occurred in the market. That makes the relatively stronger changes in the trends for low skilled more striking given the increase in wage inequality in the last 20 years. An explanation reconciling these last two facts is the subject of interest in the paper "Dynamic effects of labor supply: a mechanism explaining cross-sectional differences in hours".